Exports, Imports, and Economic Growth in Sri Lanka: Evidence from Causality and Co-integration Analysis

Abstract

This study aims to find long run causal relationship between the export and import on economic growth in Sri Lanka over the period from 1962 to 2015 using annual data. Johansen co-integration technique and vector error correction model (VECM) are used to investigate the relationships. The empirical results do not confirm a bidirectional causality between any of the variables considered but there is a unidirectional causality between export and economic growth in the short run. Further, it finds that there is a long-run equilibrium relationship between export and economic growth. Major implication of our findings is that export is matter for the economic growth of Sri Lanka than import.

Full text

5th Annual International Research Conference- 2016
Faculty of Management and Commerce- SEUSL
Exports, Imports, and Economic Growth in Sri Lanka: Evidence from
Causality and Co-integration Analysis
Riyath, MIM 1 and Jahfer, A 2
1 Department of Accountancy, Sri Lanka Institute of Advanced Technological Education
2 Department of Accountancy and Finance, South Eastern University of Sri Lanka
riyath.i@sliate.ac.lk, jahfer@seu.ac.lk
Abstract
This study aims to find long run causal relationship between the export and import on economic growth in
Sri Lanka over the period from 1962 to 2015 using annual data. Johansen co-integration technique and vector
error correction model (VECM) are used to investigate the relationships. The empirical results do not
confirm a bidirectional causality between any of the variables considered but there is a unidirectional
causality between export and economic growth in the short run. Further, it finds that there is a long-run
equilibrium relationship between export and economic growth. Major implication of our findings is that
export is matter for the economic growth of Sri Lanka than import.
Keywords: Economic growth, Export, Import, Granger causality, VECM,
1. Introduction
An important policy strategy for assessing growth
and development in developing economies is to
ascertain economic relationships between
economic growth and the trade sector for a
particular economy (Andrews, 2015). The
relationship between Export, Import and Economic
growth is an important issue among economists,
and many researchers tried to investigate this
relationship. It is widely argued that there is a two-
way causal relationship between export and
economic growth. Consequently, an extensive
empirical literature exists on the relationship
between exports and growth (Uğur, 2008). Jahfer
and Inoue (2014) find that the trade openness
affects economic growth significantly in Sri Lanka.
Hussain and Saaed (2014) state that there are four
possible propositions on a relationship between
export and economic growth: export-led growth
(ELG), growth-driven export (GDE), feedback
relationship between export and economic growth
and finally there is no relationship. Some of the
researchers found unidirectional causality and some
of them found bidirectional causality and of course
some of them could not found any evidence for
causality between export and GDP (Hussain &
Saaed, 2014).
Export-led growth (ELG) hypothesis suggest that
Export growth is often considered to be a main
determinant of the production and employment
growth of an economy. It encourages that the
overall progress of countries can be achieved not
only by mounting the quantity of manpower and
investment within the economy, but also by
increasing exports(Hussain & Saaed, 2014).
Balassa (1978), Edwards (1998) and Ramos (2001)
explain how export contribute for economic growth
of economy in four sequence of steps.
In contrast to the export-led growth hypothesis, The
growth-driven export (GDE) hypothesis suggest
that export expansion could be stimulated by
productivity gains caused by increase in domestic
levels of skilled-labor and technology (Krugman,
1984). Further it suggests that there is
unidirectional causality from economic growth to
exports but not vise-versa. Neoclassical trade
theory typically stresses the causality that runs from
home-factor endowments and productivity to the
supply of exports (Findlay, 1984). The product life
cycle hypothesis of Vernon (1966) has also
attracted significant consideration among
international trade theorists in recent years (Ramos
2001).
Lawrence and Weinstein (1999) suggest import-
lead growth (ILG) hypothesis which suggests that
economic growth could be driven primarily by
growth in imports. Endogenous growth models
show that imports can be a channel for long-ran
economic growth because it provides domestic
firms with access to needed intermediate and
foreign technology. Growth in imports can serve as
a medium for the transfer of growth-enhancing
foreign R&D knowledge from developed to
developing countries (Mazumdar, 2001).
1.1 Research Problem
Studies are find that there are relationship exist
between Export, Import and Economic growth in
Sri Lankan context. Velnampy and Achchuthan

5th Annual International Research Conference- 2016
Faculty of Management and Commerce- SEUSL
(2013) find that the export and import have the
significant positive relationship, and also, both
export and import have the significant impact on the
economic growth. Jahfer and Inoue (2014) find that
the Export and Import affects economic growth
significantly in Sri Lanka. The strong correlation of
export, import and economic growth rates has
nothing to say about a relationship between the
export import and the economic development, as it
may arise from a purely short-run relationship
(Ramos, 2001). However, It has become
meaningful when understand the casual
interrelationships among export, import and
economic growth in long run. Any research on the
long-run causal relationship among export, import
and economic growth is hard to find in Sri Lankan
context. Therefore, the test for existence of a long-
run causal relationship among export, import and
economic growth is to be performed to fill the gap
in literature in Sri Lankan context.
1.2 Research Question
What type of casual relationship exist in the long
run among the export, import and economic growth
in Sri Lanka.
1.3 Research Objective
The objective of this research is to see the
relationship among the export, import and
economic growth in Sri Lanka during the period
1960-2015.
2. Methods
In this study, we use annual data for the Sri Lanka
economy for the period 1962 to 2015 and collected
mainly from World Development Indicators
published by the World Data Bank and the reports
of Central Bank of Sri Lanka. The economic growth
is measured in terms of GDP is the sum of gross
value added by all resident producers in the
economy. Exports represent the value of all goods
and other market services provided to the rest of the
world. They include the value of merchandise,
freight, insurance, transport, travel, royalties,
license fees, and other services, such as
communication, construction, financial,
information, business, personal, and government
services. They exclude compensation of employees
and investment income. Imports represent the value
of all goods and other market services received
from the rest of the world. They include the value
of merchandise, freight, insurance, transport, travel,
royalties, license fees, and other services, such as
communication, construction, financial,
information, business, personal, and government
services. They exclude compensation of employees
and investment income and transfer payments. The
GDP, Export and Import data are measured in
constant local currency. All the data are transferred
to natural logarithms for conventional statistical
reasons.
In order to investigate the effect of export (LgX)
and import (LgM) on Economic Growth (LgY), co-
integration test and vector error correction model
(VECM) are applied. Since macroeconomic time
series data contain unit root, variables used in the
study are tested for stationary before running
causality tests. For this purpose, unit roots are tested
using augmented Dickey-Fuller (ADF) (Dickey &
Fuller, 1979) test. Furthermore, The Phillips-Perron
unit root (PP) (Phillips & Perron, 1988) test and
Dickey-Fuller (DF) are also performed due to the
possibility of the existence of structural breaks
which result in the ADF test wrongly indicating
non- stationary in what is actually a stationary
series (Hussain, 2014).
After confirming that the variables are integrated of
order one, then it is tested the existence of co-
integration relationship between the variables. The
co-integration tests are done among the variables
using the Johansen (1988) co-integration tests.
Since Johansen co-integration is sensitive to the lag
length, the Final prediction error (FPE), Akaike
information criterion (AIC), Schwarz information
criterion (SC) and Hannan-Quinn information
criterion (HQ) are used to determine the appropriate
number of lag.
3. Results and discussion
3.1 Unit root test
As a first step, to check the stationary of the
variables, the augmented Dickey-Fuller (ADF) test,
Phillips-Perron unit root (PP) and Dickey-Fuller
(DF) test were employed at level and first
difference are shown in table 1.
The lag length for the unit root tests was selected to
ensure that the residuals were white noise. It is
obvious from the unit root tests show that at level,
none of the variables represents a stationary
process. The unit root tests computed using the first
difference of GDP (LgY), exports (LgX) and
imports (LgM) indicate that all variables become
stationary at the 1% level of significance. As
differencing once produces stationary, it suggests
that each of the series GDP (LgY), the exports
(LgX) and the imports (LgM) is integrated in order
1, I(1).

5th Annual International Research Conference- 2016
Faculty of Management and Commerce- SEUSL
Table 1. Unit root test
series
Levels
1st difference
integrated
of order
DF
ADF
PP
DF
ADF
PP
LgY
0.388083
1.92251
1.866236
-5.4475
-5.64143
-5.57932
I(1)
Lag
2
0
0
0
0
0
LgX
2.035975
0.821937
1.022145
-8.04211
-8.10618
-8.06709
I(1)
Lag
0
0
0
0
0
0
LgM
1.976386
0.846284
0.818265
-5.99639
-6.66812
-6.66812
I(1)
Lag
0
0
0
0
0
0
3.2 Test for co-integration
Since all series are integrated processes of order
one. This is a necessary step in order to test the co-
integration of the variables. The co-integration tests
are done among the variables using the Johansen’s
co-integration tests to investigate long-term
equilibrium relationship among the variables.
Number of lags is selected using an optimal lag
structure in the unrestricted VAR (Table 2).
Table 2. VAR Lag Order Selection Criteria
VAR Lag Order Selection Criteria
Endogenous variables: LGY LGX LGM
Exogenous variables: C
Sample: 1962 2015
Included observations: 50
Lag
LogL
LR
FPE
AIC
SC
HQ
0
12.87059
NA
0.000135
-0.394824
-0.280102
-0.351137
1
261.0139
456.5836*
9.49e-09*
-9.960555*
-9.501669*
-9.785809*
2
268.8405
13.46180
9.99e-09
-9.913620
-9.110570
-9.607814
3
276.5584
12.34869
1.06e-08
-9.862337
-8.715123
-9.425472
4
286.5274
14.75405
1.04e-08
-9.901095
-8.409717
-9.333170
* indicates lag order selected by the criterion
LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion
Johansen’s approach derives two likelihood
estimators for the co-integration rank: a trace test
and a maximum Eigen value test. Table 3 presents
summarized co-integration results between the
variables. Co-integration results indicate the
existence of long-run association between GDP,
Export and import in Sri Lanka. Therefore, VECM
can be used to investigate the relationships among
the selected variables.
Table 3. Co-integration Test
Hypothesized
No. of CE(s)
Trace Test
Maximum Eigenvalue Test
Eigenvalue
Trace
Statistic
Critical
value at 5%
Prob.**
Trace
Statistic
Critical
value at 5%
Prob.**
None *
0.479482
45.2289
24.2759
0.0000
33.95237
17.7973
0.0001
At most 1
0.155796
11.2765
12.3209
0.0743
8.806796
11.2248
0.1291
At most 2
0.046385
2.46974
4.12990
0.1371
2.469744
4.129906
0.1371
Trace test and Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
3.3 Granger-causality in the Vector error
correction model
The number of co-integrating relationships found in
Table 3 will result in a corresponding number of
residual series, and hence error correction terms

5th Annual International Research Conference- 2016
Faculty of Management and Commerce- SEUSL
(ECT) to be used in the subsequent vector error
correction model VECM.
To test whether export and import Granger-causes
GDP, the following VECM is estimated.
∆ = 10 + �11 ∆−

=1
+ �12 ∆−

=1
+ �13 ∆−

=1
+ 14 − + 
To test whether GDP and import Granger-causes
Export, the following VECM is estimated.
∆ = 20 + �21 ∆−

=1
+ �22 ∆−

=1
+ �23 ∆−

=1
+ 24 − + 
To test whether GDP and export Granger-causes
Import, the following VECM is estimated.
∆ = 30 + �31 ∆−

=1
+ �32 ∆−

=1
+ �33 ∆−

=1
+ 34 − + 
Since the variables are co-integrated, there are long-
term relationships among the variables under
consideration. Table 4 presents summary results of
VECM with respect to Economic growth, export
and import under the three models. The estimated
ECT is negative and highly significant in model 1.
The result is supporting the co-integration among
the variables represented by model. The result
suggest there is a long run causality from export and
import on economic growth. Furthermore, the
export coefficient is positive and significant in the
model 1. It suggests that the export causes GDP in
the short run. However, the import is not causes
GDP in the short run in Sri Lankan Economy.
According to VECM test result of the model 2 and
3, the error correction terms are insignificant.
Therefore, there is no causality from economic
growth and import on export in the long run. In
addition, there is no causality from economic
growth and export on import in the long run.
Table 4. VECM test
Dependent Variable
D(LGY)
D(LGY)
Coefficient
Std. Error
t-Statistic
Prob.
ECT
-0.01408
0.002516
-5.59382
0.0000
D(LGY(-1))
0.186353
0.152814
1.21948
0.2286
D(LGX(-1))
-0.10969
0.044285
-2.47695
0.0168
D(LGM(-1))
0.067499
0.035634
1.894194
0.0642
Dependent Variable
D(LGX)
D(LGX)
Coefficient
Std. Error
t-Statistic
Prob.
ECT
-0.01245
0.009276
-1.34162
0.1860
D(LGY(-1))
0.10193
0.56329
0.180954
0.8572
D(LGX(-1))
-0.21311
0.163242
-1.30549
0.1979
D(LGM(-1))
0.096829
0.131353
0.737166
0.4646
Dependent Variable
D(LGM)
D(LGM)
Coefficient
Std. Error
t-Statistic
Prob.
ECT
-0.0034
0.012458
-0.27265
0.7863
D(LGY(-1))
0.709439
0.756542
0.937738
0.3531
D(LGX(-1))
0.040643
0.219246
0.185377
0.8537
D(LGM(-1))
-0.02456
0.176417
-0.13919
0.8899

5th Annual International Research Conference- 2016
Faculty of Management and Commerce- SEUSL
The results support the one-way causal relationship
between economic growth and export and there is
no significant causality between import and export
in Sri Lankan Economy. The result support Export
led hypothesis rather than Growth-driven exports.
4. Summary and conclusions
This study investigates the long run causal
relationship among export, import and economic
growth in Sri Lanka using annual data over the
period 1962 to 2015. The stationary of the data is
tested using ADF, DF and PP test. Johansen co-
integration technique and the VECM are used to
estimate the effect of export and import on
economic growth. Johansen co-integration test
finds that export, import and economic growth are
co-integrated. VECM results demonstrate that there
is a long-run equilibrium relationship among the
variables and a unidirectional causality between the
export and economic growth in the short run.
Further, no strong evidence that import causes
economic growth of Sri Lanka. Moreover, major
implication of our findings is that export is matter
for the economic growth of Sri Lanka than import.
Refernces
Andrews, A. P. (2015). eExports, Imports, and
Economic Growth in Liberia: Evidence
from Causality and Cointegration
Analysis. Journal of Management Policy
and Practice, 16(3), 95.
Balassa, B. (1978). Exports and economic growth:
further evidence. Journal of development
Economics, 5(2), 181-189.
Dickey, D. A., & Fuller, W. A. (1979). Distribution
of the estimators for autoregressive time
series with a unit root. Journal of the
American statistical association, 74(366a),
427-431.
Edwards, S. (1998). Openness, productivity and
growth: what do we really know? The
economic journal, 108(447), 383-398.
Findlay, R. (1984). Growth and development in
trade models. Handbook of international
economics, 1, 185-236.
Hussain, M. A. (2014). Economic Growth, Exports
and Imports in Pakistan: Granger Causality
Analysis. The Journal of Business in
Developing Nation, 13.
Hussain, M. A., & Saaed, A. A. (2014).
Relationship between Exports, Imports,
and Economic Growth in Saudi Arabia:
1990-2011. Evidence from Co-Integration
and Granger Causality Analysis. Journal
of Emerging Trends in Economics and
Management Sciences, 5(3), 364.
Jahfer, A., & Inoue, T. (2014). Financial
development, foreign direct investment
and economic growth in Sri Lanka.
International Journal of Economic Policy
in Emerging Economies, 7(1), 77-93.
Johansen, S. (1988). Statistical analysis of
cointegration vectors. Journal of economic
dynamics and control, 12(2), 231-254.
Krugman, P. (1984). Import protection as export
promotion: International competition in
the presence of oligopoly and economies
of scale. Monopolistic competition and
international trade, 180-193.
Lawrence, R. Z., & Weinstein, D. E. (1999). Trade
and growth: import-led or export-led?
Evidence from Japan and Korea: National
bureau of economic research.
Mazumdar, J. (2001). Imported machinery and
growth in LDCs. Journal of development
Economics, 65(1), 209-224.
Phillips, P. C., & Perron, P. (1988). Testing for a
unit root in time series regression.
Biometrika, 75(2), 335-346.
Ramos, F. F. R. (2001). Exports, imports, and
economic growth in Portugal: evidence
from causality and cointegration analysis.
Economic Modelling, 18(4), 613-623.
Uğur, A. (2008). Import and economic growth in
Turkey: Evidence from multivariate VAR
analysis. Journal of economics and
Business, 11(1-2), 54-75.
Velnampy, T., & Achchuthan, S. (2013). Export,
Import, and Economic Growth: Evidence
from Sri Lanka. Journal of Economics and
Sustainable Development, 4(9), 147-155.
Vernon, R. (1966). International investment and
international trade in the product cycle.
The quarterly journal of economics, 190-
207.